JP3226580B2 - Image processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to at least binarizing image data obtained by raster reading with a predetermined threshold value, removing isolated point pixels from a binarized image signal, and The present invention relates to an image processing apparatus having a function of converting the resolution of data in the sub-scanning direction to １ ／.

[0002]

2. Description of the Related Art Generally, when image data obtained by raster reading is converted into an image signal of a black-and-white binary image, processing for improving characteristics of an image reading system or removing noise is performed. . Further, for example, pseudo halftone reading processing using a dither matrix is performed so that halftones can be expressed to some extent in the output image. Further, for example, as in a group 3 facsimile machine, the reading line density in the sub-scanning direction is 3.85 (lines / mm) (hereinafter referred to as standard line density) and 7.7 (lines / mm) (hereinafter high lines). Density), the reading system always reads at a high linear density, and when a standard linear density is designated, an image signal for two lines is read. A line density conversion process for combining into one line is performed.

FIG. 12 shows a conventional example of a control system of a document reading apparatus applied to a group 3 facsimile apparatus.

In FIG. 1, an analog image signal AV raster-scanned by a document reading system (not shown) and output from an image input unit 1 is converted by an analog / digital converter 2 into a corresponding digital image signal DV. The digital image signal DV is applied to a fixed binarization processing section 3 and a halftone binarization processing section 4.

[0005] The fixed binarization processing section 3 outputs a digital image signal D
A predetermined threshold value is applied to V to form a fixed binarized image signal SV. The output signal is applied to one input terminal A of the selector 5 and one input terminal of the image area discriminating unit 6. ing.

The halftone binarization processing unit 4 forms a halftone binarized image signal HV by periodically applying a predetermined threshold matrix to the digital image signal DV. 5 and the other input terminal of the image area determination unit 6.

When a halftone reading operation is designated by a halftone reading command signal HT output from a higher-level control unit (not shown), the image area determining unit 6 outputs a fixed binarized image signal SV and a halftoned binary signal. Based on the image signal HV, it is determined whether or not the target pixel belongs to the character image area. When it is determined that the target pixel belongs to the character image area, the selection signal SL is set to the logical H level. increase. This selection signal SL is applied to a control signal input terminal of the selector 5. Note that the image area determination unit 6 outputs the halftone reading command signal H
If the halftone reading operation is not specified by T,
The selection signal SL is fixed at the logic H level.

When the selection signal SL is at a logic H level, the selector 5 selects the fixed binary image signal SV applied to the input terminal A, while the selection signal SL is at a logic L level. When it is, the halftone binary image signal HV applied to the input terminal B is selected, and the selected signal is output to the isolated point removing unit 7 and the buffer memory 8 as the selected image signal LV.

The buffer memory 8 has a capacity capable of storing the selected image signal LV for two lines, the selected image signal LVa one line before the input selected image signal LV and the selection two lines before the selected image signal LV. The image signal LVb is supplied to the isolated point removing unit 7.
Output to

The isolated point removing unit 7 outputs a halftone reading command signal H
When the halftone reading operation is not specified by T, FIG.
When it is detected that the contents of the 3 × 3 pixel matrix shown in FIG. 3A match the states shown in FIGS. 3B and 3D, and when the respective states are detected, As shown in FIGS. 3C and 3E, the pixel state of the target pixel X located at the center of the pixel matrix is inverted to remove an isolated point. The corrected image signal CV is applied to the OR processing unit 9 and the line buffer 10. When the halftone reading operation is designated by the halftone reading command signal HT, the isolated point removing unit 7 outputs the input selected image signal L
V is output as it is as the corrected image signal CV.

The line buffer 10 has a capacity capable of storing the corrected image signal CV for one line, and the corrected image signal CVa one line before the input corrected image signal CV.
Is output to the OR processing unit 9.

The OR processing unit 9 specifies a standard linear density by a linear density command signal STD output from a higher-level control unit (not shown), and specifies a halftone reading operation by a halftone reading command signal HT. If not, the correction image signal CV and the correction image signal CVa to be input are subjected to a logical OR process on a pixel-by-pixel basis, and the processing result is output as a binary image signal BW. In addition, the OR processing unit 9 determines that the high linear density is specified by the linear density command signal STD.
Alternatively, when the halftone reading operation is designated by the halftone reading command signal HT, the input correction image signal CV
Is output directly as a binary image signal BW. In this case, the logical H level corresponds to the black pixel level.

In the above configuration, when the halftone reading is designated by the standard resolution, the halftone reading operation is designated by the halftone command signal HT, and the standard resolution is designated by the line density command signal STD.

Thus, the image area determination unit 6 determines the value of the fixed binarized image signal SV and the halftone binarized image signal HV.
It is determined whether or not the target pixel belongs to the character area, and the logical level of the selection signal SL is set based on the determination result.

In this case, the isolated point removing unit 7
The input selected image signal LV is output as it is as the corrected image signal CV, and the OR processing unit 9 outputs the input corrected image signal CV as it is as the binary image signal BW.

Therefore, in this case, the selected image signal LV selected by the selector 5 is output as the binarized image signal BW.

When the non-halftone reading is designated by the standard resolution, the non-halftone reading operation is designated by the halftone command signal HT, and the standard resolution is designated by the line density command signal STD.

As a result, the image area determination section 6 outputs the selection signal S
Since L is fixed at the logical H level, the selected image signal LV output from the selector 5 matches the fixed binarized image signal SV output from the fixed binarization processing unit 3.

The isolated point removing unit 7 executes the above-described isolated point removing operation based on the selected image signal LV, the selected image signal LVa one line before and the selected image signal LVb two lines before, and outputs the result. It is output as a corrected image signal CV.

The OR processing unit 9 performs a logical OR process on the corrected image signal CV and the corrected image signal CVa one line before, for each pixel, and outputs the processing result as a binary image signal BW.

When halftone reading is designated at high resolution, a halftone reading operation is designated by a halftone command signal HT, and a high resolution is designated by a line density command signal STD.

Thus, the image area determination section 6 determines the value based on the fixed binarized image signal SV and the halftone binarized image signal HV.
It is determined whether or not the target pixel belongs to the character area, and the logical level of the selection signal SL is set based on the determination result.

In this case, the isolated point removing unit 7
The input selected image signal LV is output as it is as the corrected image signal CV, and the OR processing unit 9 outputs the input corrected image signal CV as it is as the binary image signal BW.

When non-halftone reading is designated at high resolution, non-halftone reading operation is designated by the halftone command signal HT, and high resolution is designated by the line density command signal STD.

As a result, the image area determination section 6 outputs the selection signal S
Since L is fixed at the logical H level, the selected image signal LV output from the selector 5 matches the fixed binarized image signal SV output from the fixed binarization processing unit 3.

In this case, the isolated point removing unit 7
Based on the selected image signal LV, the selected image signal LVa one line before, and the selected image signal LVb two lines before, the above-described isolated point removal operation is performed, and the result is output as a corrected image signal CV. In this case, the OR processing unit 9
The input corrected image signal CV is output as it is as a binarized image signal BW.

As described above, in the conventional apparatus, when the standard resolution is set and the halftone reading is set, the image area is determined by the image area determining section 6 and the isolated point removing section 7 and the OR The operation of the processing unit 9 does not work effectively. When the standard resolution and the non-halftone reading are set, the image area determination is not performed, and the isolated point removing operation and the OR processing are performed.

When high resolution and halftone reading are set, the image area is determined by the image area determining section 6 and the operations of the isolated point removing section 7 and the OR processing section 9 are effectively performed. Does not work. When high resolution and non-halftone reading are set, the image area determination is not performed, the isolated point removing operation is performed, and the OR processing is not performed.

As described above, conventionally, when the isolated point removal processing and the OR processing are applied to the halftone image, there is a disadvantage that the image quality of the halftone read image is deteriorated. The isolated point removal processing and the OR processing are not applied to the halftone image. For a non-halftone image, a fixed binary image signal is always selected, and an isolated point removal process and an OR process are applied.

[0030]

However, such a conventional apparatus has the following inconveniences.

That is, the isolated point removal processing and the OR processing are not applied to the halftone image.
Therefore, even for a non-halftone image portion of an image in which a halftone image and a non-halftone image coexist, neither the isolated point removal process nor the OR process is applied, and the image quality of such a portion cannot be improved. Further, the image area determination unit 6 includes a buffer memory for holding an image signal necessary for the processing, separately from the isolated point removal unit 7 and the OR processing unit 9, so that the circuit scale of the control system is large. This hinders miniaturization and cost reduction of the device.

The present invention has been made in view of such circumstances, and can improve the image quality of an image in which a halftone image and a non-halftone image are mixed, and reduce the size and cost of a control system for a read image. It is an object to provide an image processing device that can be realized.

[0033]

According to the present invention, there is provided a fixed binarizing means for binarizing image data obtained by raster reading with a predetermined threshold value, and an image signal outputted from the fixed binarizing means. On the other hand, in an image processing apparatus having a function of removing an isolated point pixel and converting the resolution of image data in the sub-scanning direction to 、, the image signal output from the fixed binarizing means And an output from the fixed binarization means at a timing shifted by one line from the pixel of interest processed by the first isolated point removing means. A second isolated point removing unit that determines whether the image signal to be output is an isolated point pixel and removes the isolated point; and outputs of the first isolated point removing unit and the second isolated point removing unit. Sub-scanning of image signal by calculating logical sum of signals Those having a logical sum processing means for converting the resolution to 1/2.

The image processing apparatus further comprises pseudo halftone processing means for performing pseudo halftone processing on image data obtained by raster reading, and fixed binarization means for binarizing the image data with a predetermined threshold value. If the image data is included in the non-character area, the image signal output from the pseudo halftone processing means is selected. In an image processing apparatus having a function of selecting a signal, removing isolated point pixels, and converting the resolution of image data in the sub-scanning direction to 複数, a plurality of lines output from a fixed binarization unit. Image area determining means for determining whether the pixel of interest is included in a character area or a non-character area based on the minute image signal,
A first method for determining whether or not the image signal output from the fixed binarization means is an isolated point pixel and removing the isolated point
And whether the image signal output from the fixed binarizing means at a timing shifted by one line from the pixel of interest processed by the first isolated point removing means is an isolated point pixel. And an OR operation of output signals of the first isolated point removing means and the second isolated point removing means for removing the isolated points, and calculating the logical sum of the image signals in the sub-scanning direction. A logical sum processing means for converting the resolution to 1/2, and an output signal of the logical sum processing means when the image area determining means determines the character area, while the image area determining means selects the non-character area. When the judgment is made, a signal selecting means for selecting an output signal of the pseudo halftone means is provided. Further, the first isolated point removing means and the second isolated point removing means take out required image signals from the image signals for a plurality of lines held by the image area determining means, respectively, and perform isolated point removing processing. I do it.

[0035]

Therefore, since the processing results of the first isolated point removing means and the second isolated point removing means are subjected to the logical sum processing, no special buffer memory is required for the logical sum processing means. Thus, the amount of memory required for the device can be reduced.
Further, any one of the image signal as a result of the OR processing after removal of the isolated point and the image signal as a result of pseudo halftone reading is
Since the selection is made based on the image area determination result, the image quality of an image in which a halftone image and a non-halftone image are mixed can be improved.

[0036]

Embodiments of the present invention will be described below in detail with reference to the accompanying drawings.

FIG. 1 shows a group 3 facsimile apparatus according to one embodiment of the present invention.

In the figure, a control unit 11 performs control processing of each unit of the facsimile apparatus and facsimile transmission control procedure processing.
Stores a control processing program to be executed by the control unit 11, various data necessary for executing the processing program, and the like, and configures a work area of the control unit 11. The parameter memory 13 This is for storing various information unique to the group 3 facsimile apparatus.

The scanner 14 has a resolution of 8 in the main scanning direction.
(Dots / mm) and the resolution in the sub-scanning direction is 3.
Two types of reading resolutions, a standard resolution of 85 (lines / mm) and a high resolution of 8 (dots / mm) in the main scanning direction and 7.7 (lines / mm) in the sub-scanning direction The plotter 15 is for recording and outputting the image at one of the standard resolution and the high resolution, and the operation display unit 16 is provided for the facsimile apparatus. , And comprises various operation keys and various display devices.

The encoding / decoding section 17 encodes and compresses the image signal and decodes the encoded and compressed image information into the original image signal.
Is for storing a large number of image information in a coded and compressed state.

The group 3 facsimile modem 19 is for realizing a group 3 facsimile modem function. The group 3 facsimile modem 19 has a low-speed modem function (V.21 modem) for exchanging transmission procedure signals and mainly exchanges image information. High-speed modem function (V.29 modem,
V. 27 ter modem).

The network control device 20 is for connecting the facsimile apparatus to a public telephone line network and has an automatic transmission / reception function.

The control unit 11 and the system memory 1
2, parameter memory 13, scanner 14, plotter 1
5, operation display unit 16, encoding / decoding unit 17, image storage device 18, group 3 facsimile modem 19, and
The network control device 20 is connected to a system bus 21, and data exchange between these components is mainly performed through the system bus 21.

Data exchange between the network controller 20 and the group 3 facsimile modem 19 is directly performed.

FIG. 2 shows a control system of the scanner 14 according to one embodiment of the present invention. In this figure, the same parts as those in FIG. 12 and the corresponding parts are denoted by the same reference numerals.

In the figure, a digital image signal DV output from the analog / digital converter 2 is an MTF (spatial frequency) correction operation unit 25, a line having a capacity capable of storing two lines of digital image signals DV. It is added to a buffer 26 and a halftone binarization processing unit 27, respectively. Also, the digital image signal DV delayed by one line from the input digital image signal DV is output from the line buffer 26.
A digital image signal DV2 delayed by 1 and 2 lines is output, and these digital image signals DV1 and DV2 are M
TF correction calculator 25 and halftone binarization processor 27
Has been added to each.

The MTF correction operation section 25 performs MTF correction operation processing for correcting image blur caused by the reading optical system, and the correction result is sent to the image area separation section 28 as a corrected digital image signal DVc. Have been added. This MTF correction calculation process is performed, for example, as shown in FIG.
The pixel level of the target pixel X located at the center of the pixel matrix of the × 3 size is corrected based on the following equation (I).

X ′ = (8X− (2B + D + E + 2G)) / 2 (I)

Here, X ′ indicates the level of the target pixel X after correction, X indicates the level of the target pixel X before correction,
B, D, E, and G indicate the levels of the peripheral pixels B, D, E, and G, respectively.

The image area separation unit 28 separates a character and a picture (photograph and halftone image) on a white background into pixels based on the input corrected digital image signal DVc, and designates a target pixel to be processed as a character. When the area is determined as an area, the determination signal SK is set to a logical H level (active value), and when the pixel of interest is determined to be a non-character area, the determination signal SK is set to a logical L level (inactive value). The determination signal SK is applied to a control signal input terminal of the selector 29.

Further, the image area separating section 28 has a threshold value THb for determining the input corrected digital image signal DVc as a black pixel and a threshold value T for determining the input digital image signal DVc as a white pixel.
Hw, each of which is ternarized. Based on the ternarized signals, character separation processing for determining contrast, continuity of white pixels, and continuity of black pixels, and extraction of a white pixel region to extract a white background Is performed. For this process, a black pixel signal (binary signal) determined as a black pixel is held for four lines, and a white pixel signal (binary signal) determined as a white pixel is held. Of the corrected digital image signal DVc, the black pixel signal BK1 one line before the input corrected digital image signal DVc is applied to the isolated point removing unit 30, and the black pixel signal B two lines before
K2 is added to the isolated point removing unit 30 and the isolated point removing unit 31, and the black pixel signal BK3 three lines before is output to the isolated point removing unit 30.
The black pixel signal BK4 four lines before is applied to the isolated point removing unit 31. Here, the target pixel to be processed by the image area separation unit 28 appears in the black pixel signal BK3.

The isolated point removing unit 30 applies the same isolated point removing operation as the isolated point removing unit 7 of FIG. 12 to the pixel (pixel X1 of FIG. 4) one line before the pixel of interest of the image area separating unit 28. The processing result is applied to one input terminal of the OR circuit 32 as a corrected binary signal BWp. The isolated point removing unit 31 applies the isolated point removing operation to the pixel of interest (the pixel X2 in FIG. 4) of the image area separating unit 28, and the processing result is ORed as the corrected binary signal BWc. It is applied to the other input of the circuit 32. In this case, the logic H level is set to a value representing a black pixel.

When the standard resolution is specified by the resolution signal CL applied from the control unit 1, the OR circuit 32 performs a logical OR operation on the input corrected binary image signal BWp and corrected binary image signal BWc in pixel units. And outputs the processed result as an output binary image signal BWo.
When the high resolution is designated by the resolution signal CL, the corrected binary image signal BWc is output to the output terminal A of the selector 29 as the output binary image signal BWo.

The halftone binarization processing section 27 applies a halftone binarization process based on the input digital image signals DV, DV1 and DV2, and the processing result is a halftone binarization process. The image signal HT is supplied to the buffer memory 33. The buffer memory 33 stores, for the pixel of interest,
This is for matching the output timing of the halftone image signal HT with the output timing of the output binarized image signal BWo output from the OR circuit 32. The output halftone image signal output from the buffer memory 33 is used. HTo is applied to the input terminal B of the selector 29.

The selector 29 selects the input terminal A when the determination signal SK is at the logic H level, outputs the output binary image signal BWo to the external device as the binary image signal BW, and outputs the determination signal When SK is at the logic L level, the input terminal B is selected, and the output halftone binarized image signal H is output.
To is output to an external device as a binary image signal BW.

FIG. 5 shows an example of the halftone binarization processing section 27.

In the same figure, the smoothing processing unit 35 has a digital image signal DV, which is one more than the digital image signal DV.
The digital image signal DV1 before the line and the digital image signal DV2 two lines before the digital image signal DV are added, and the smoothing processing unit 35 performs the processing based on these digital image signals DV, DV1, and DV2. For the pixel of interest X located at the center of the same 3 × 3 pixel matrix as 3, the following equation (II) is executed to cause the pixel of interest X to include components of surrounding pixels.

X ′ = (2E + (B + D + F + H) / 2) / 4 (II)

The smoothing image signal DVs output from the smoothing processing unit 35 is applied to the dither binarization processing unit 3
6 has been added. The dither binarization processing unit 36 periodically extracts the elements of the dither threshold matrix stored in the threshold storage unit 37, and binarizes the smoothed image signal DVs based on the value of the dither threshold matrix. The processing result is output as a halftone binary image signal HT.

FIG. 6 shows an example of the image area separating section 28.

In the figure, the corrected digital image signal DVc
Is added to the binarization processing unit 41 and the binarization processing unit 42. The binarization processing unit 41 performs a binarization process on the corrected digital image signal DVc with a threshold THb for determining a black pixel, and the processing result is a black pixel signal BK
And a line buffer 44 having a capacity capable of storing the black pixel signals BK for four lines.

From the line buffer 44, the black pixel signal B
A black pixel signal BK1 one line before K, a black pixel signal BK2 two lines before, a black pixel signal BK3 three lines before, and a black pixel signal BK4 four lines before are output. Is a black pattern matching processing unit 4
3 and the black pattern matching processing unit 45,
The black pixel signal BK2 is supplied to the black pattern matching processing unit 43,
The black pixel signal BK3 is applied to the black pattern matching processing unit 45 and the black pattern matching processing unit 46, and the black pixel signal BK4 is applied to the black pattern matching processing unit 46. Has been added to These black pixel signals BK1, BK2, BK3, and BK4 are output to the isolated point removing units 30 and 31.

The black pattern matching processing units 43, 45,
46 performs the following pattern matching processing based on the input black pixel signals for three lines. That is, a 3 × 3 pixel matrix centered on the pixel of interest is extracted, and the content of the pixel matrix matches any of FIGS. 7A to 7H and FIG.
If it matches any of (a) to (f), the target pixel is determined as a black active pixel, and the output signals BA1, BA2, and BA3 are set to active values. Here, FIGS. 7A to 7H and FIGS.
In (f), the symbol “○” represents a pixel whose black pixel signal value is an inactive value, the symbol “●” represents a pixel whose black pixel signal value is an active value,
The symbol “x” represents a pixel whose black pixel signal value may be either an inactive value or an active value, and the symbol “▲” represents a pixel at least one of which has a black pixel signal value of an active value.

Therefore, signals BA1, BA2, BA3
Indicate whether the pixel at the position of the target pixel on the current line (target pixel line) two lines before, one line before, is a black active pixel, and the signals BA1, BA2, BA3 is added to the black area determination unit 47.

The black area determination section 47 outputs signals BA1, BA
2. Based on BA3, a signal of a 3 × 3 pixel matrix centering on the pixel of interest is extracted, and it is checked whether there are two or more pixels set to the active value among the extracted signals. When there are two or more pixels set to the active value, the output signal BB is set to the active value. This signal BB is supplied to the AND processing unit 48
Is applied to one input end.

The binarization processing section 42 binarizes the corrected digital image signal DVc with a threshold value THw for judging a black pixel. The processing result is a black pattern as a white pixel signal WT. A matching processing unit 49, and
It is added to a line buffer 50 having a capacity capable of storing the white pixel signals WT for five lines.

From the line buffer 50, the white pixel signal W
White pixel signal WT1 one line before T, white pixel signal WT2 two lines before, white pixel signal WT3 four lines before
The white pixel signal WT4 before the line and the white pixel signal WT4 four lines before are output. The white pixel signal WT1 is added to the white pattern matching processing unit 49, the white pattern matching processing unit 51, and the white background pixel detection unit 52. The white pixel signal WT2 is applied to a white pattern matching processing unit 49, a white pattern matching processing unit 51, a white pattern matching processing unit 53, and a white background pixel detection unit 52, and the white pixel signal WT3 is applied to the white pattern matching processing unit 51, Pattern matching processing section 53 and white background pixel detection section 52
The white pixel signal WT4 is applied to a white pattern matching processing unit 53 and a white background pixel detection unit 52, and the white pixel signal WT5 is applied to the white background pixel detection unit 52.

The white pattern matching processing sections 49, 51,
53 performs the following pattern matching processing based on the input white pixel signals for three lines. That is, a 3 × 3 pixel matrix centered on the target pixel is extracted, and if the contents of the pixel matrix match any of FIGS. 9A to 9F, the target pixel is set to white. The output signal is determined as an active pixel, and the output signals WA1, WA2, and WA3 are set to active values.

Therefore, signals WA1, WA2, WA3
Indicate whether the pixel at the position of the target pixel on the current line (target pixel line) two lines before, one line before, is a white active pixel, and the signals BA1, BA2, BA3 is added to the white area determination unit 54.

The white area determination section 54 outputs the signals WA1 and WA
2. Based on WA3, a signal of a 3 × 3 pixel matrix centering on the target pixel is extracted, and it is checked whether there are two or more pixels set to the active value among the extracted signals. If there is one or more pixels set to the active value, the output signal WW is set to the active value. This signal WW is output to the AND processing unit 48
Is applied to the other input terminal.

When both the signal BB and the signal WW are set to the active value, the AND processing unit 48 sets the output signal CC to the active value.
The signal CC is output to the area determination unit 55 and the signal CC
It is added to a line buffer 56 having a storage capacity for lines. From the line buffer 56, the signal CC 1
The signal CC1 before the line, the signal CC2 two lines before, and the signal CC3 three lines before are output. These signals CC1, CC2, and CC3 are applied to the area determination unit 55.

The area judging section 55 extracts a 5 × 4 pixel matrix signal as shown in FIG. 10 based on the inputted signals CC, CC1, CC2 and CC3 for four lines, and When one or more pixels having an active value are detected, the target pixel X (signal CC)
It is determined that the active pixel is the active pixel, and the value of the output signal DD is set to the active value. This signal DD is applied to one input terminal of the character area determination section 57.

The white background pixel detector 52 receives the input white pixel signals WT, WT1, WT2, WT3, WT for five lines.
4, a 5 × 5 pixel matrix shown in FIG. 11A is extracted based on WT5, and if all 10 consecutive pixels in the pixel matrix are white pixels, the pixel matrix is located at the center of the pixel matrix. The target pixel X is determined to be a white background pixel, and the output signal WK is set to an active value.

That is, in the pixel matrix, the pixels A to J in the first and second rows are all white pixels, or the pixels F to N in the second and third rows are all white pixels, or
Pixels K to S in the row and the fourth row are all white pixels, or
Pixels O to Y in the fourth and fifth rows are all white pixels, or pixels A to U in the first and second columns are all white pixels,
The pixels B to V in the third and third columns are all white pixels, or
When the pixels C to W in the third and fourth columns are all white pixels, or when the pixels D to Y in the fourth and fifth columns are all white pixels, such a block of white pixel area is detected. , The target pixel X is determined as the active pixel.

The signal WK output from the white background pixel detection section 52 is supplied to an area determination section 58 and a line buffer 59 having a storage capacity capable of storing two lines of signals WK.
Has been added to The line buffer 59 outputs a signal WK1 one line before the signal WK and a signal WK2 two lines before the signal WK. These signals WK1 and WK2 are
It is added to the area determination unit 58.

The area judging section 58 extracts a signal of a 3 × 1 size pixel matrix as shown in FIG. 11B based on the input signals WK, WK1 and WK2 for three lines, and When it is detected that there is one or more active pixels, the target pixel X at the center is determined as the active pixel, and the value of the output signal WL1 is set to the active value. This signal WL1 is applied to the area determination unit 60.

The area determining unit 60 determines whether the 1 × 1 line signal WL1 shown in FIG.
A signal of a 32-pixel matrix is extracted, and when it is detected that there is one or more active pixels in the pixel matrix, the target pixel X at the end is determined as an active pixel, and the value of the output signal WL2 is determined. To the active value. This signal WL2 is applied to the other input terminal of the character area determination section 57.

The character area determination section 57 sets the value of the determination signal SK to the active value when the signal DD is at the active value and the signal WL2 is at the active value. Set the value of SK to the inactive value. This determination signal SK is applied to the selector 29.

As described above, the character separation function of the image area separation unit 28 according to the present embodiment uses the corrected digital image signal DV.
The contrast is determined by binarizing c, and further,
A region where both a black continuous pixel and a white continuous pixel exist in the local region is separated as a character region pixel.

Therefore, the photographic image, the halftone dot image, and the character image are determined as follows by the character separating function.

First, a photographic image is basically a flat image. After ternarization, since a black pixel and a white pixel do not exist very close to each other, it is determined as a non-character area. On the other hand, as described above, since the image area separation unit 28 according to the present embodiment determines the continuity of black pixels and the continuity of white pixels, the halftone dots are dispersedly arranged. In the halftone dot image, the high screen ruling halftone, the halftone highlight part, and the halftone shadow part are determined as the non-character areas. In the character image, the periphery of the edge of the character is determined as a character area.

On the other hand, the white area separation function of the image area separation section 28 detects a cluster of white pixels and expands the area to the periphery to detect a white area including a character portion.

Therefore, the photographic image, the halftone dot image, and the character image are determined as follows by the white background separating function.

First, in the photographic image and the halftone image, since a cluster of white background pixels is detected, a halftone portion and a shadow portion are determined as a non-white background region, and a highlight portion is determined as a white background region. . However, at the border of the highlight,
In both the halftone portion and the shadow portion, a portion corresponding to the expanded size is determined as a white background region. In a character image, a white background area including a character portion can be detected.

Further, the area included in the area determined as the character area by the character separation function and the area included in the area determined as the white background area by the white background separation function is determined as a character area by the character area determination unit 57. Therefore, the image area separation unit 28 according to the present embodiment can efficiently separate the character area from the non-character area.

With the above configuration, when the reading operation of the scanner 4 is started, the analog image signal AV is sequentially output from the image input unit 1 and converted into the corresponding digital image signal DV by the analog / digital converter 2. , MTF correction calculation unit 25, line buffer 26, and halftone binarization processing unit 27.

The necessary data is stored in the line buffer 26, and the digital image signal DV
1 and DV2 are sequentially output, the MTF correction calculation unit 25
Is started, and the corrected digital image signal DVc is sequentially output to the image area separation unit 28. At the same time, the halftone binarization processing section 27 performs the above-described halftone binarization processing using the smoothing process and the dither threshold, and the resulting halftone binarized image signal HT is sequentially stored in a buffer memory. 33.

As described above, the image area separation unit 28 executes the character separation function and the white background separation function, and delays the time corresponding to three lines and four pixels from the input timing of the corrected digital image signal DVc. , A determination signal SK corresponding to the pixel of interest to be processed is output and applied to the selector 29.

On the other hand, the black pixel signals BK1, BK2, BK3, and BK4 are output from the image area separating section 28, and the isolated point removing section 31 outputs the corrected binary image signal BWc of the target pixel. At the same time, the isolated point removing unit 30 outputs the corrected binary image signal BWp corresponding to the pixel located at the target pixel after one line.

At this time, when the standard resolution is specified by the control unit 1 and the value of the resolution signal CL is a value corresponding to the standard resolution, the OR processing unit 32 outputs the corrected binary image signal BWc and the corrected binary image signal BWc. The binarized image signal BWp is subjected to a logical OR process on a pixel basis, and the processing result is output to the selector 29 as an output binarized image signal BWo.

On the other hand, the halftone binarized image signal HT output from the halftone binarization processing section 27 is supplied to the buffer memory 3 at a timing when a time corresponding to three lines and four pixels has elapsed.
3 as the output halftone binary image signal HTo
9 is output.

Therefore, the output binarized image signal BWo and the output halftone binarized image signal HTo relating to the pixel at the same position on the read original are simultaneously applied to the selector 29, and the selector 29 is also supplied with the output binarized image signal HTo. , A determination signal SK corresponding to the same pixel is output and applied to the selector 29.

Therefore, the selector 29 outputs the judgment signal S
When K is an active value (logic H level) representing a character area, the output binarized image signal BWo is output as a binarized image signal BW, and the determination signal SK is output.
Is an inactive value (logic L level) representing a non-character area, the output halftone binary image signal H
To is output as a binary image signal BW.

Thus, in this case, the isolated point removing unit 3
Since the output binarized image signal BWo in a state where the isolated points are removed at 0 and 31 is output from the OR processing unit 32, the image of the standard resolution character area also has higher image quality and higher quality than the isolated points removed. Become.

At the same time, when the output pixel of interest is included in the character area, the output binary image signal BWo is output as the binary image signal BW, and the output pixel of interest is output to the non-character area. , The output halftone binarized image signal HTo is output as a binarized image signal, thereby improving the image quality of a standard resolution image in which a character image and a halftone image are mixed.

On the other hand, when the high resolution is designated by the control section 1 and the value of the resolution signal CL is a value corresponding to the high resolution, the OR processing section 32 outputs the corrected binary image signal B
Wc is output as an output binary image signal BWo, and the selector 29 outputs the output binary image signal BWo output from the OR processing unit 32 when the determination signal SK is at the active value. BW, and when the determination signal SK has an inactive value, the output halftone binary image signal H output from the buffer memory 33 is output.
To is output as a binary image signal BW. Therefore,
The image quality of a high-resolution image is improved.

As described above, in this embodiment, since the signals required by the isolated point removing units 30 and 31 are obtained from the image area separating unit 28, the buffer memory necessary for the isolated point removing units 30 and 31 is provided. This can be omitted, and the cost of the apparatus can be reduced. In addition, since the processing results of the two isolated point removing units 30 and 31 are input to the OR processing unit 32, the buffer memory required for the OR processing unit 32 can be omitted, and the apparatus cost can be reduced. Also in the image area separation unit 28, the line buffer that can be shared is shared, and the data necessary for the processing of the black area determination unit 47 and the white area determination unit 54 is 3
Are formed by the three black pattern matching processing units 43, 45, 46 and the white pattern matching processing units 49, 51, 53. Therefore, the buffer memory is omitted in the black region determination unit 47 and the white region determination unit 54. ,
Equipment costs can be reduced.

After the isolated point removal processing for two lines is performed by the isolated point removal units 30 and 31, the OR processing unit 32 performs a logical sum operation on the processing results.
The isolated point removal processing can be applied to the standard resolution image as in the case of the high resolution image, and the image quality of the image in the character area is improved.

Further, since the output of the OR processing unit 32 and the output of the halftone binarization processing unit 27 are selected by the selector 29, the character of the standard resolution image is the same as that of the high resolution image. In the area, the image signal obtained by the fixed binarization processing can be obtained, and in the non-character area, the image signal obtained by the halftone reading processing can be obtained. The image quality of the image can be greatly improved.

In the above-described embodiment, the present invention is applied to the group 3 facsimile apparatus.
The present invention can be similarly applied to an apparatus having a similar document reading function.

[0101]

As described above, according to the present invention,
Since the processing results of the first isolated point removing means and the second isolated point removing means are ORed, the OR processing means does not require a special buffer memory, thereby providing a necessary device. The amount of memory can be reduced. In addition, since one of the image signal obtained as a result of OR processing and the image signal obtained as a result of pseudo halftone reading is selected based on the image area determination result, the halftone image and The effect of improving the image quality of an image in which non-halftone images are mixed can be obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a group 3 facsimile apparatus according to one embodiment of the present invention.

FIG. 2 is a block diagram illustrating an example of a control system of the scanner.

FIG. 3 is a schematic diagram showing an example of a 3 × 3 pixel matrix;

FIG. 4 is a diagram showing 3 × 3 pixels applied by two isolated point removing units;
FIG. 3 is an explanatory diagram for describing a pixel matrix of a size.

FIG. 5 is a block diagram illustrating an example of a halftone binarization processing unit;

FIG. 6 is a block diagram illustrating an example of an image area separation unit.

FIG. 7 is an explanatory diagram showing a part of a pattern determined by a black pattern matching processing unit.

FIG. 8 is an explanatory diagram showing another part of the pattern determined by the black pattern matching processing unit.

FIG. 9 is an explanatory diagram showing a pattern determined by a white pattern matching processing unit.

FIG. 10 is an explanatory diagram showing an example of a 5 × 4 size pixel matrix.

FIG. 11: 5 × 5 size, 1 × 3 size, and 32
FIG. 3 is an explanatory diagram illustrating an example of a × 1 pixel matrix.

FIG. 12 is a block diagram illustrating a conventional example of an image processing apparatus.

FIG. 13 is an explanatory diagram for explaining isolated point removal processing.

[Explanation of symbols]

 25 MTF (spatial frequency) correction operation unit 26, 44, 50, 56, 59 line buffer 27 halftone binarization processing unit 28 image area separation unit 29 selector 30, 31 isolated point removal unit 32 OR processing unit 33 buffer memory 41 , 42 Binarization processing unit 43, 45, 46 Black pattern matching processing unit 47 Black region determination unit 48 AND processing unit 49, 51, 53 White pattern matching processing unit 52 White background pixel detection unit 54 White region determination unit 55, 59, 60 area determination section 57 character area determination section

────────────────────────────────────────────────── ─── Continuation of front page (56) References JP-A-2-131668 (JP, A) JP-A-3-117171 (JP, A) JP-A-5-73672 (JP, A) JP-A-5-67267 94522 (JP, A) JP-A-3-278767 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H04N 1/38-1/393 H04N 1/40-1/409 H04N 1/46 H04N 1/60

Claims (3)

(57) [Claims] 1. A fixed binarization unit for binarizing image data obtained by raster reading with a predetermined threshold value, and removing an isolated point pixel from an image signal output from the fixed binarization unit. And the resolution of the image data in the sub-scanning direction is reduced by half.
A first isolated point removing unit that determines whether an image signal output from the fixed binarizing unit is an isolated point pixel and removes the isolated point, It is determined whether or not the image signal output from the fixed binarizing means is an isolated point pixel at a timing shifted by one line from the pixel of interest processed by the first isolated point removing means, and the isolated point is removed. 2 and a logical sum of the output signals of the first isolated point removing means and the second isolated point removing means to convert the resolution of the image signal in the sub-scanning direction to 1/2. An image processing apparatus comprising a logical sum processing means. 2. A method according to claim 1, further comprising pseudo halftone processing means for performing pseudo halftone processing on image data obtained by raster reading, and fixed binarization means for binarizing the image data with a predetermined threshold value. If the image data is included in the non-character area, the image signal output from the pseudo halftone processing means is selected. In an image processing apparatus having a function of selecting a signal, removing isolated point pixels, and converting the resolution of image data in the sub-scanning direction to 複数, a plurality of lines output from a fixed binarization unit. Image area determining means for determining whether the pixel of interest is included in the character area or non-character area based on the image signal of the minute, and the image signal output from the fixed binarizing means is an isolated point pixel Judge whether it is And an image signal output from the fixed binarizing means at a timing shifted by one line from a pixel of interest processed by the first isolated point removing means. A second isolated point removing means for determining whether the pixel is a pixel and removing an isolated point, and a logical sum of output signals of the first isolated point removing means and the second isolated point removing means are calculated. OR processing means for converting the resolution of the image signal in the sub-scanning direction to て,
When the image area determining means determines the character area, the output signal of the logical sum processing means is selected. When the image area determining means determines the non-character area, the output signal of the pseudo halftone means is selected. An image processing apparatus comprising a signal selecting means for selecting. 3. The first isolated point removing means and the second isolated point removing means respectively extract necessary image signals from a plurality of lines of image signals held by the image area determining means, and 3. A removing process is performed.
The image processing apparatus according to any one of the preceding claims.